Analysis and quantitation of the carbohydrates in honey using high

Department of Applied Microbiology and Food Science, 149 John Mitchell Building,. University of Saskatchewan, Saskatoon, Saskatchewan, Canada S7N 0W0...
0 downloads 0 Views 544KB Size
J. Agric. Food Chem. 1990, 38, 1828-1832

1828

Analysis and Quantitation of the Carbohydrates in Honey Using High-Performance Liquid Chromatography Kevin W. Swallow and Nicholas H. Low' Department of Applied Microbiology and Food Science, 149 John Mitchell Building, University of Saskatchewan,Saskatoon, Saskatchewan,Canada S7N OW0

Carbohydraseenzymes present in honey catalyze transglucosylation reactions which result in the formation of over 20 structurally similar oligosaccharides. These oligosaccharides make up a small percentage of the total carbohydrates in honey. The combination of these two facts makes the analysis and quantitation of these minor oligosaccharides difficult. Methodology was developed using highperformance liquid chromatography to qualitatively and quantitatively analyze 20 of the minor oligosaccharides present in four unifloral Canadian honeys.

INTRODUCTION

Honey is one of the most complex mixtures of carbohydrates produced in nature. The major carbohydrates in honey are glucose and fructose, which account for 6575 7% of the total soluble solids in honey and 85-95 7% of honey carbohydrates (White, 1979). The remaining carbohydrates are a mixture of at least 11 disaccharides (Siddiqui and Furgala, 1967; Siddiqui, 1970; Low and Sporns, 1988), 11trisaccharides (White and Hoban, 1959; Siddiqui and Furgala, 1968), and several higher oligosaccharides. Several methods have been attempted to analyze the totalcarbohydrate profile of honey. These include a variety of chromatographic methods such as paper chromatography (Partridge, 1949), thin-layer chromatography (Tate and Bishop, 1962), gas-liquid chromatography (GLC) (Echigo, 1970), and high-performance liquid chromatography (HPLC) [for a review, see Honda (1984)l. All of these methods can be utilized to separate and quantitate the major carbohydrates in honey; however, the analysis of the more complex minor carbohydrates has proven much more difficult. The method of choice for the analysis of structurally similar carbohydrates such as those present in honey is GLC. GLC analysis of carbohydrates is very popular due to the high sensitivity obtained by the flame ionization detector (FID). With FID, detection limits for oligosaccharides are in the order of 40 ppb (Low and Sporns, 1988). There are, however, two main drawbacks with the use of GLC to analyze carbohydrates: (1)The sugars have to be in a single tautomeric form or multiple peaks will occur from one sugar. (2) Since carbohydrates are not volatile, they must be derivatized before GLC analysis. The most common method of derivatizing carbohydrates for GLC analysis was developed by Sweeley et al. (1963) using trimethylchlorosilane and hexamethyldisilazane. Due to these problems, the use of HPLC for carbohydrate analysis has received a great deal of attention. Relatively little sample preparation is required, and the carbohydrates can be isolated after separation for further analysis. Gel permeation (Whistler and Anisuzzaman, 1980), reverse-phase (McGinnis et al., 1986), and silica (Linden and Lawhead, 1975) columns have been used for HPLC analysis of carbohydrates, but the most popular are amino-bonded silica (Truong Van Den et al., 1986) and ~

~~~

~

~~

* Author to whom correspondence should be addressed. 0021-8561 19011438-1828$02.50/0

fixed-ion resin (Thomas and Lobel, 1976) columns. The amino-bonded silica columns use acetonitrile/water as the mobile phase and are probably the most common resin used today for the separation of carbohydrates. However, these stationary phases suffer from poor resolution of structurally similar carbohydrates and loss of amino groups due to cleavage of the silyl ether bonds (Shaw, 1988). The fixed-ion resins use water as the mobile phase, but also suffer from poor resolution of structurally similar carbohydrates. In addition, these stationary phases are less efficient unless elevated column temperatures are used and are highly compressible due to low cross-linking. Separation and detection of the minor oligosaccharides in honey by HPLC is difficult for two main reasons. The first problem is due to oligosaccharides with structural similarities. Siddiqui and Furgala (1967,1968)found that the oligosaccharides present in honey comprised mainly disaccharides which are either glucose-glucose or glucosefructose linked. The other major problem is that the concentration of the minor oligosaccharides in honey is low, making detection difficult. To some degree, detection problems have been alleviated by the development of new detectors for HPLC, such as mass detectors (MacRae and Dick, 1981) and electrochemical detectors (Hughes and Johnson, 1982), and design improvements in refractive index (RI) detectors. However, detection limit problems still exist which cannot be alleviated by simply concentrating the honey sample. Increasing the concentration of honey not only increases the concentration of these oligosaccharides but also results in a dramatic increase in glucose and fructose which swamp the active sites in the column, inhibiting oligosaccharide analysis. A relatively new method of carbohydrate separation was introduced by Rocklin and Pohl (1983) using anionexchange chromatography. By use of this method of separation, carbohydrates are ionized in a high pH eluant and separation occurs due to variations in pK, values of the carbohydrates in the alkaline mobile phase. These stationary phases achieve remarkably good resolution of structurally similar carbohydrates using simple aqueous buffers. In this paper, we introduce methodology employing HPLC to separate 20 structurally similar carbohydrates. Using anion-exchange chromatography in conjunction with a pulsed amperometric detector, we qualitatively and quantitatively analyzed the minor oligosaccharides present in four honeys of known botanical origin. 8 1990 American Chemical Society

J. Agric.

HPLC of Honey Carbohydrates

Food Chem., Vol. 38, No. 9, 1990 1829

Table I. Retention Time (T,), Response Factor (RF),and Weight Percent of the Carbohydrates in Alfalfa, Alsike, Canola, and Trefoil Honey SamDles (Average of Three Replicates) wt%

peak 1 2 3 4 5b

6 7 8'

9 10

11 12 13d 14 15 16 17 18

19 20 d

carbohydrate neotrehalose glucose fructose melibiose isomaltose maltulose sucrose kojibiose turanose gentiobiose palatinose melezitose isomaltotriose nigerose maltose 1-kestose theanderose laminaribiose isopanose erlose panose maltotriose laminaritriose

T,,min

RF 1.19 2.55 2.56 1.07 0.52

Alfalfa trace 33.9 f 0.4 37.0 f 0.4 NDa 0.30 f 0.01

Alsike trace 35.6 f 0.4 38.5 f 0.4 ND 0.31 f 0.01

Canola trace 40.0 f 0.4 36.2 f 0.4 ND 0.12 f 0.01

Trefoil trace 33.2 f 0.4 37.1 f 0.4 ND 0.35 f 0.01

1.18 f 0.09 0.36 f 0.01

0.90 f 0.04 0.36 f 0.05

0.046 f 0.008 0.37 f 0.05

0.55 f 0.03 0.49 f 0.02

39.3 f 0.4 43.2 f 0.5 47.6 f 0.5 51.2 f 0.4 53.2 f 0.5

1.09 0.92 1.04 4.29 3.49 0.93 1.06 2.14 1.32

0.067 f 0.001 0.038 f 0.002 0.038 f 0.002 0.30 f 0.01 1.18 f 0.05

trace 0.056 f 0.003 0.028 f 0.005 0.34 f 0.02 1.00 f 0.02

trace 0.041 f 0.001 trace 0.23 f 0.01 0.76 f 0.05

trace 0.047 f 0.001 trace 0.40 f 0.01 1.11f 0.04

55.4 f 0.6 57.2 f 0.7 60.3 f 0.5 61.4 f 0.4 63.9 f 0.5 66.4 f 0.6 71.1 f 0.5

0.92 3.10 1.23 0.63 0.86 1.35 3.10

0.38 f 0.04 0.10 f 0.01 ND 3.43 f 0.20 0.080 f 0.003 0.10 f 0.01 trace

0.34 f 0.01 trace 0.10 f 0.01 3.97 f 0.22 trace 0.087 f 0.011 ND

0.067 f 0.011 0.053 f 0.002 ND 0.26 f 0.01 0.025 f 0.002 0.023 f 0.001 ND

0.103 f 0.001 0.132 i 0.005 trace 2.86 f 0.20 0.090 f 0.003 0.059 f 0.004 ND

7.8 f 0.3 13.7 f 0.1 15.5 f 0.1 19.9 f 0.1 24.9 f 0.3 28.2 f 0.4 30.0 f 0.4 37.9 f 0.4

a ND, not detected. Calculated as isomaltose. Concentrationof turanose and gentiobiose varies in honey and therefore cannot be calculated. Calculated as maltose.

MATERIALS AND METHODS

Table 11. Relative Percentages (w/w) of Glucose, Fructose, Total Oligosaccharides and HzO in Alfalfa, Alsike, Canola, and Trefoil Honey Samples

The honey carbohydrate standards were obtained from the honey % glucose0 76 fructoseo % oligosaccharides % HzOa following sources: D-glucose and sucrose (a-D-glucopyranosyl 0-D-fructofuranoside) were obtained from BDH Chemicals. 7.5 19.6 f 0.6 Alfalfa 33.9 f 0.4 37.0 f 0.4 D-Fructose was obtained from Aldrich Chemical Co., Inc. Ko7.5 17.1 f 0.6 Alsike 35.6 f 0.4 38.5 f 0.4 jibiose [ O-a-~-glucopyranosyl-( 1-+2)-~-glucopyranose]was ob2.0 18.9 f 0.6 Canola 40.0 f 0.4 36.2 f 0.4 tained from Koch-Light Laboratories Ltd. Neotrehalose (a+ 21.9 f 0.6 6.2 Trefoil 33.2 f 0.4 37.1 f 0.4 glucopyranosyl P-D-glucopyranoside) was a gift from Dr. I. R. a Average of three replicates. Siddiqui, Agriculture Canada. Laminaribiose [0-0-D-glucopyranosyl-(1-3)-~-glucopyranose] and laminaritriose [0-@-D- igin of each honey was confirmed by pollen analysis (Low et al., glucopyranosyl-(1-3)-0-@-~-glucopyranosyl-( 1+3)-~-glu1989). copyranose] were gifts from Dr. E. Reese of the U S . Army Sample Preparation. In preparation for oligosaccharide Natick Research and Development Laboratories. Maltulose [0analysis, 1 g of each honey was dissolved in 19.0 g of HPLC a-~-glucopyranosyl-(l-+4)-~-fructose] and nigerose [O-WDgrade HzO and passed through a (2-18 Sep-Pak cartridge (Waglucopyranosyl-(1-3)-~glucopyranose]were synthesized (Hicks ters Associates) and 3 cm3 of AG 501-X8 mixed bed resin, 20et ai., 1983) in our laboratory. Palatinose [o-a-D-glucopyrano50 mesh (Bio-Rad Laboratories). The samples were then passed syl-(l-.6)-~fructose], gentiobiose [O-B-D-glucopyranosyl-(1+6)through 3 cm3 of AG 1-X4anion-exchange resin, 100-200 mesh, glucopyranose], isomaltose [ 0-a-D-glucopyranosyl-(1+6)-D(Bio-Rad) to remove organic acids. The monosaccharides were glucopyranose], melibiose [ O-a-~-galactopyranosyl-(l-6)-~- removed from the honey by a procedure modified from Whisglucopyranose], maltose [ O-a-~-glucopyranosy1-(1-4)-~- tler and Durso (1950). The samples were stirred with 4.0 g of glucopyranose], turanose [O-a-~-glucopyranosy1-(1~3)-~-activated charcoal, 50-200 mesh (Fisher Scientific Co.) for 17 h fructose], maltotriose [ O-a-~-glucopyranosy1-(1-4)-0-a-~- at 4 "C. After mixing, the samples were placed on a 3.0 cm diglucopyranosyl-(1-4)-~-glucopyranose], isomaltotriose [O - ~ - D ameter column containing 4.0 g of 50/50 (w/w) mix of activatglucopyranosyl-(1-6)-0-a-~-glucopyranosyl-(l-.6)-~-glu- ed charcoal and Celite (Fisher). Approximately 99% of the copyranose], and melezitose [0-a-D-glucopyranosyl-(1-3)-0-@monosaccharides were removed from the column by washing ~-fructofuranosyl-(2-1) a-~-glucopyranoside]were obtained with 1 L of 0.1% (v/v) ethanol at room temperature a t a flow from Sigma Chemical Co. I-Kestose [ 0-a-D-glucopyranosylrate of 10.0 mL/min. The remaining oligosaccharides were (l-2)-j3-~-fructofuranosyl- (1-2) @-~-fructofuranoside] , erlose eluted from the column with 500 mL of a 60 "C solution of 50% [0-a-Dglucopyranosyl-(1-4)-a-~glucopyranoyl@-~fructofura(v/v) ethanol at the same flow rate. The filtrate was frozen a t noside], and theanderose [ O-a-D-glucopyranosyl-(1-6)-a-D-70 "C and dried at 30 "C in a Lab Con Co freeze dryer (Freeze glucopyranosyl @-~-fructofuranoside] were gifts from Dr. S. Dry 5). Ten milliliters of deionized water was added to the lyoChiba, Department of Agriculture, Hokkaido University. Paphilized honey samples. The samples were stored a t -20 "C unnose [0-a-Dglucopyranosyl-(1+6)-0-a-t+glucopyranosyl-(1-+4)til required for analysis. Sample preparation for glucose and ~-glucopyranose]and isopanose [O-cY-n-glucopyranosyl-( 1-4)fructose quantitation was achieved by simple dilution of each O-a-~-glucopyranosyl-(1+6)-~-glucopyranose]were synthesized honey with deionized water. All samples were passed through (Wolfram and Koizumi, 1967) in our laboratory. a 0.2-rm nylon 66 filter (Rainin Instrument Co.) to remove parAlfalfa, Alsike, Canola, and Trefoil honey samples were obticulate matter. Ninhydrin and Bradford tests were carried out tained in the following manner. Four-frame nuclei with 1 kg of on the samples to ensure that no amino acids or proteins were bees, a queen, 1frame of brood and 3 empty combs were placed present. in the middle of a field with a minimum of 8 ha (20 acres). All To ensure that no loss of oligosaccharides had occurred durthe floral sources were in full bloom during the honey collecing the removal of glucose and fructose, a solution containing sution. The frames were collected 7-10 days later, and the honcrose, maltose, and maltotriose was placed on the charcoal/ Celite column. The column was washed with 0.1% ethanol at ey was extracted by using a 2-frame hand extractor. Samples were stored at -20 "C until required for analysis. Botanical orambient temperature and 50% ethanol a t 60 "C as described

Swallow and Low

1830 J. Agric. Food Chem., Vol. 38,No. 9, 1990 P

i

111

TIME i m i n l

Figure 1. HPLC chromatogram of the 23 honey carbohydrate standards (see Table I for a list of the carbohydrates). earlier, and both eluates were saved. HPLC analysis of the eluates revealed that greater than 99% of the sucrose, maltose, and maltotriose was recovered during the charcoal/Celite treatment. HPLC Analysis of the Carbohydrates in Honey. The resulting samples were analyzed on a Dionex Bio LC 4000 gradient HPLC containing a 5 0 - ~ L sample loop. Separation of the carbohydrates was carried out on two Dionex 10-rm Carbo Pac PA1 pellicular anion-exchange columns (4 X 250 mm) connected in series. The flow rate was 0.70 mL/min, and the carbohydrates were detected by a PAD (pulsed amperometric detector) with a gold electrode and triple pulsed amperometry at a sensitivity of 10K. The electrode was maintained at the following potentials and durations: E1 = 0.05 V (tl 120 ms); Et = 0.80 V ( t 2 = 120 ms); Eg = -0.60 V ( t 3 = 420 ms). A postcolumn delivery system of 0.3 M sodium hydroxide (NaOH) at a flow rate of 0.80 mL/min was used to prevent base-line drift. The following gradient elution was used to achieve separation of the oligosaccharides: 0.1 M NaOH for 4 min; after 20 min, the mobile phase was 0.1 M NaOH and 0.03 M sodium acetate (NaOAc); after 50 min, the mobile phase was 0.1 M NaOH and 0.1 M NaOAc. This eluant was held for 10 min before a 0.3 M NaOH wash was used to remove the acetate ions from the columns. Following the wash step, the columns were reequilibrated with 0.1 M NaOH in preparation for the next injection. The carbohydrates eluting from the columns were plotted by a Spectra Physics Model 4290 integrator. Triplicate injections of all samples were carried out to ensure correct integration of the carbohydrates. RESULTS AND DISCUSSION The vast majority of nectars contain exclusively the sugars glucose, fructose, and sucrose (Baker and Baker, 1983). Research has shown that the minor oligosaccharides in honey arise from the transglucosylation activity of a-and &glucosidases contributed by honeybees (White and Maher, 1953; Low e t al., 1986). These carbohydrases transfer glucose moieties to other carbohydrates during the hydrolysis of sucrose, resulting in the formation of many complex oligosaccharides. Twenty-three carbohydrates previously identified in honey (Siddiqui and Furgala, 1967,1968; Siddiqui, 1970; Low and Sporns, 1988) were pooled and analyzed by HPLC. The HPLC chromatogram of these standard carbohydrates is shown in Figure 1. By use of our methodology, 20 of the 23 honey carbohydrate standards were resolved. Each standard was injected separately, and the standard mixture was spiked with each individual carbohydrate to determine the response fador and confirm the identity of each oligosaccharide. Coelution of the carbohydrates isomaltose/maltulose (peak 5), turanose/ gentiobiose (peak 81, and maltose/l-kestose (peak 13) occurred during the optimal conditions developed for oligosaccharide separation. Table I lists the retention time (Tr),response factor (RF), and weight percent of the carbohydrates identified in each

b

8'-

30

~

~

-

TIME h n I

Figure 2. HPLC chromatogram of the carbohydrates in Alfalfa honey (see Table I for a list of the carbohydrates).

=I

0

80

30 TIME imin 1

Figure 3. HPLC chromatogram of the carbohydrates in Alsike honey (see Table I for a list of the carbohydrates). of the honey samples analyzed. The concentration of the carbohydrates was determined by using response factors calculated from the carbohydrate standards. Estimation of the concentration of isomaltose (peak 5) and maltose (peak 13) in honey can be achieved since the amount of maltulose and 1-kestose in honey is small relative to isomaltose and maltose (Low and Sporns, 1988; Siddiqui and Furgala, 1967, 1968). Quantitation of either turanose or gentiobiose (peak 8) could not be accomplished with this elution program. The glucose and fructose concentration for each honey was determined by HPLC employing untreated (without charcoal/Celite chromatography) honey samples (Table 11). In addition, the moisture content of each honey was determined by using the Karl Fischer titration method. The results from these analyses fall within the range of the 490 floral honeys analyzed by White et al. (1962). Figures 2-5 are HPLC chromatograms of the carbohydrates in Alfalfa, Alsike, Canola, and Trefoil honey samples, respectively, following removal of >99% of the

J. Agric. Food Chem., Vol. 38, No. 9, 1990

HPLC of Honey Carbohydrates

1831

tigating the oligosaccharide fingerprint profiles for a number of food products to test this hypothesis. ACKNOWLEDGMENT

We thank the following collaborators: Peter Sporns, University of Alberta, Edmonton, Alberta, Canada, for providing helpful suggestions during the preparation of the manuscript; Agriculture Canada, for providing financial support during the research (K.S.). LITERATURE CITED Baker, H. G.; Baker, I. A Brief Historical Review of the Chemistry of Floral Nectar. In The Biology of Nectaries; Columbia University Press: New York, 1983;pp 126-152. Echigo, T. Determination of Sugars in Nectar and Honey, and the Mutarotation Coefficient of Glucose, by Gas-Liquid Chromatography. Bull. Fac. Agric. Tamagawa Uniu. 1970,10,

3-12. 0

ea

30 TIME lmin 1

Figure 4. HPLC chromatogram of the carbohydrates in Canola honey (see Table I for a list of the carbohydrates).

Hicks, K. B.; Symanski, E. V.; Pfeffer, P. E. Synthesis and HighPerformance Liquid Chromatography of Maltulose and Cellobiose. Carbohydr. Res. 1983,112,37-50. Honda, S. High-Performance Liquid Chromatography of Moncand Oligosaccharides. Anal. Biochem. 1984,140, 1-47. Hughes, S.;Johnson, D. C. High-Performance Liquid Chromatographic Separation with Triple-Pulse Amperometric Detection of Carbohydrates in Beverages. J. Agric. Food

Chem. 1982,30,712-714. Linden, J. C.; Lawhead, C. L. Liquid Chromatography of Saccharides. J. Chromatogr. 1975,105,125-133. Low N. H.; Sporns, P. Analysis and Quantitation of Minor Diand Trisaccharides in Honey, Using Capillary Gas Chromatography. J. Food Sci. 1988,53 (2),558-561. Low, N. H.; Va Vong, K.; Sporns, P. A New Enzyme, 0-Glucosidase, in Honey. J. Apic. Res. 1986,25 (3),178-181. Low, N. H.; Nelson, D. L.; Sporns, P. Carbohydrate Analysis of Western Canadian Honeys and Their Nectar Sources to Determine the Origin of Honey Oligosaccharides. J. Apic. Res.

1988,27(4),245-251. Low, N. H.; Schweger, C.; Sporns, P. Precautions Using Melissopalynology (Pollen Analysis). J. Apic. Res. 1989,in press. MacRae, R.; Dick, J. Analysis of Carbohydrates Using the Mass Detector. J. Chromatogr. 1981,210,138-145. McGinnis, G. D.; Prince, S.; Lowrimore, J. The Use of ReversePhme Columns for Separation of Unsubstituted Carbohydrates.

scn

J. Carbohydr. Chem. 1986,5,83-97. b

eo

30 TIME

ImnI

Figure 5. HPLC chromatogram of the carbohydrates in Trefoil honey (see Table I for a list of the carbohydrates).

glucose and fructose by charcoal/Celite chromatography. Although relative oligosaccharide concentration varies from one honey sample to the next, the overall oligosaccharide pattern does not differ significantly. Therefore, these oligosaccharide patterns could be used as “fingerprints” for honey authenticity. The largest variation in honey oligosaccharides is that found in Canola honey (Figure 4). Canola flowers contain very little sucrose (Low et al.,1983, which results in a significantly different carbohydrate profile. One example is an almost total absence of erlose (peak 17) which predominates in other unifloral honeys. Concentration of the oligosaccharides using charcoal/ Celite chromatography followed by HPLC analysis with anion-exchange chromatography in conjunction with a pulsed amperometric detection system affords the qualitative and quantitative analyses of structurally similar oligosaccharides. This methodology can be used to determine the minor oligosaccharides present in a variety of foods. These patterns can then be used as fingerprints for the authenticity of foods. We are currently inves-

Partridge, S. M. Aniline Hydrogen Phthalate as a Spraying Reagent for Chromatography of Sugars. Nature 1949,164,443. Rocklin, R. D.; Pohl, C. A. Determination of Carbohydrates by Anion Exchange Chromatography with Pulsed Amperometric Detection. J. Liq. Chromatogr. 1983,6,1577-1590. Shaw, P. E. Handbook of Sugar Separations in Foods by HPLC; CRC Press: Boca Raton, FL, 1988;pp 31-42. Siddiqui, I. R. The Sugars of Honey. Adu. Carbohydr. Chem.

Biochem. 1970,25,285-309. Siddiqui, I. R.; Furgala, B. Isolation and Characterization of Oligosaccharides from Honey. Part I. Disaccharides. J. Apic. Res. 1967,6 (3),139-145. Siddiqui, I. R.; Furgala, B. Isolation and Characterization of Oligosaccharides from Honey. Part 11. Trisaccharides. J. Apic.

Res. 1968, 7 (l),51-59. Sweeley, C. C.; Bentley, R.; Makita, M.; Wells, W. W. GasLiquid Chromatographyof TrimethylsilylDerivatives of Sugars and Related Substances. J. Am. Chem. SOC.1963,85,2497-

2507. Tate, M. E.; Bishop, C. T. Thin Layer Chromatography of Carbohydrate Acetates. Can. J . Chem. 1962,40,1043-1048. Thomas, J.; Lobel, L. H. Chromatographic Separation of Glucose and Mannose on Cation-ExchangeResin. Anal. Biochem. 1976, 73,222-226. Truong Van Den; Biermann, C. J.; Marlett, J. A. Simple Sugars, Oligosaccharides and Starch Concentrationsin Raw and Cooked Sweet Potato. J. Agric. Food Chem. 1986,34,421-425.

1832 J. Agric. FoodChem., Vol. 38, No. 9, 1990

Whistler, R. L.; Anisuzzaman, A. K. M. Gel Permeation Chromatography. Methods Carbohydr. Chem. 1980, 8, 4549. Whistler, R. L.; Durso, D. F. Chromatographic Separation of Sugars on Charcoal. J. Am. Chem. SOC.1950, 72,677-679. White, J. W., Jr. Composition of Honey. In H o n e y : A Comprehensive survey; Crane, E., Ed.; Heinemann in Cooperation with International Bee Research Association: London, 1979. White, J. W., Jr.; Hoban, N. Composition of Honey. IV. Identification of the Disaccharides. Arch. Biochem. Biophys. 1959,80, 386-392. White, J. W., Jr.; Maher, J. Transglucosidation by Honey Invertase. Arch. Biochem. Biophys. 1953, 42 (2), 360-367. White, J. W., Jr.; Riethof, M. L.; Subers, M. H.; Kushnir, I.

Swallow and Low

Compition of American Honeys. Tech. Bull.-US. Dep. Agric. 1962, 1261. Wolfram, M. L.; Koijumi, K. A Chemical Synthesis of Panose and an Isomeric Trisaccharide. J . Org. Chem. 1967,32,656660. Received for review December 29,1989. Accepted April 11,1990. Registry No. Neotrehalose,58591-1; glucose, 50-99-7;fructose, 57-48-7; melibiose, 585-99-9; isomaltose, 499-40-1; sucrose, 5750-1; kojibiose, 2140-29-6; turanose, 547-25-1;palatinose, 13718 94-0;melezitose,597-12-6;isomaltotriose, 3371-50-4;nigerose, 49748-3; maltose, 69-79-4; theanderose, 21291-36-1; laminaribiose, 34980-39-7; isopanose, 32581-33-2; erlose, 13101-54-7; panose, 33401-87-5; maltotriose, 1109-28-0; laminaritriose, 3256-04-0; maltulose, 17606-72-3;gentiobiose, 554-91-6; 1-kestose, 470-699.